Lotka-Volterra模型於台資銀行大陸分行營收分析

Abstract

本研究係以Lotka-Volterra模型探討台資銀行赴大陸投資發展策略，比較官股銀行以及非官股銀行至大陸投資的動態競合關係。其中官股銀行定義有至大陸設分行且曾為台灣政府百分之百持股之銀行，其餘定義為非官股銀行。本研究係以台資銀行赴中國大陸設立分行之利息收入以及手續費收入測試官股銀行以及非關股銀行使否存在著競爭或獵食與獵物者的關係，並且對於官股以及非官股銀行預測未來關係是否存在著均衡，最後，除了本研究所提出的Lotka-Volterra模型進行預測，還會運用Bass模型進行預測，且比較兩模型何者較能精確的預測台資銀行赴中國大陸投資。 本研究結果顯示出，在利息收入部分，官股銀行會幫助非官股銀行成長，但是非官股銀行並不會幫助官股銀行成長。其原因可能為銀行業利息收入主要來源為企業放款，而官股銀行較早進入大陸市場佈局從事放款，取得利息收入後，激勵非官股銀行有信心跟進在中國設分行進行貸款，或擴大貸款授信額度。在手續費收入部分，而台資銀行在中國設立分行所收的手續費，包括承兌手續費以及保管手續費，官股與非官股銀行手續費互相增長的情況，代表台灣官股銀行在中國大陸分行從事的服務愈多，手續費越高，會促成台灣非官股銀行在中國大陸分行從事的服務愈多，手續費因而提升。自兩岸金融開放以來，台資銀行持續在中國設立分行，本研究以Lotka-Volterra模型探究官股銀行以及非官股銀行赴中國利息收入與手續費收入間的關係，提供海峽兩岸政府金融管理的參考。本研究均衡分析的結果顯示在利息收入方面，官股銀行和非官股銀行會收斂到333.2和122.6，官股銀行的均衡解約為非官股銀行的三倍。而手續費收入方面，官股銀行和非官股銀行會收斂到26.2和6.1，官股銀行的均衡解約為非官股銀行的四倍，說明官股銀行和非官股銀行大陸分行在長期競合下，將會達到穩定均衡的狀態。精確度分析的結果顯示本研究所採用的Lotka-Volterra模型準確度比Bass模型準確度來得精確。

This study employs Lotka-Volterra model to analyze the competitive operational relations of Chinese branches between government-owned-banks and non-government-owned-banks. Two pairs of Lotka-Volterra equations are estimated in this work. One pair is the interest revenues of government-owned-banks and non-government-owned-banks. The other pair is the commission revenues of government-owned-banks and non-government-owned-banks from the Chinese branches of Taiwanese Banks. The estimation parameter result shows that the interest revenues from government-owned-banks will enhance those of non-government-owned-banks. This implies that the early entry of Taiwanese government-owned-banks into the mainland China. After government-owned-banks set up branches in China and offer loans to receive interest revenues, non-government-owned-banks follow government-owned-banks to set up branches in China and earn interest revenues. On the other hand, the results of parameter estimation show the commission revenues of government-owned-bank enhance the commission revenues of non-government-owned-bank, and vice versa. Regarding model’s forecast accuracy, the Lotka-Volterra model is more accurate than the Bass model in predicting interest revenues and commission revenues. In the long run, both the interest and commission revenues of government-owned-banks and non-government-owned-banks will reach a stable equilibrium. This study contributes to Taiwan’s and China’s governments the effective methods to measure the competition and cooperation in banking industry.